INDEX

Explanations

locations or geographical references related to London

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

onies

-0.84

hemy

-0.83

icio

-0.82

utation

-0.73

mble

-0.72

utations

-0.70

earance

-0.70

ONY

-0.70

merce

-0.70

lic

-0.69

POSITIVE LOGITS

 Underground

1.08

 borough

0.94

 Calling

0.91

 Heath

0.91

 Bridge

0.90

 Spit

0.89

 Borough

0.88

 Harbour

0.88

 Metropolitan

0.87

 Beckham

0.86

Activations Density 0.025%

locations or geographical references related to London

No Comments

No Known Activations